1. Field of the Invention
The present invention relates generally to intravascular devices. More particularly, it concerns self-expandable woven intravascular devices for use as stents, occluders or filters, the methods of making the same, and the apparatus and methods for delivery of the same into a living creature.
2. Description of Related Art
Intravascular devices that serve as stents or filters constructed using a plain weave, such as the stent disclosed in U.S. Pat. No. 4,655,771 to Wallsten (hereinafter, the WALLSTENT), have a propensity to show a high-degree of elongation axially with diameter reduction. This is especially significant, when the angle of the crossing wires is close to the largest possible. The closer that the angle between the wires is to 180°, the more the corresponding elongation of the stent is at a given percentage of decrease in diameter. Any discrepancy between the diameters of the stent and the vessel can result in a considerable elongation of the stent. Simultaneously, the woven type stent has the largest expansile force and hence the biggest resistance to outer compression when the angle between the crossing wires is close to 180°. In some applications, such as outer compression by a space occupying lesion, the increased radial force may be advantageous. The disadvantage of a propensity for elongation is that great care must be taken when delivering such a stent in a vessel or non-vascular tubular structure in order to properly position it.
A further disadvantage of intravascular devices formed using a plain weave, is that they are often incapable of maintaining their shape when bent. For example, when such a stent is being delivered through a tortuous passageway with many turns, upon being bent, the weave of the stent tightens (e.g., the angle of the crossing wires approaches 180°). As a result of this tightening, the diameter of the stent increases and the length of the stent decreases. Consequently, the diameter of the stent may exceed the diameter of the vessel or structure through which it is traveling, impeding the delivery of the stent or causing the stent to lodge in the vessel. This problem may be due in part to the use of weave materials such as stainless steel, which exhibit poor shape memory. This problem may also be due to the free, unclosed wires used to form the stent. The free sharp ends can create potential complications by penetrating, or perforating the wall of the tubular structure where such a stent is placed. Further, steps that have been taken to eliminate the free, sharp ends, such as connection with U-shaped members using welding, glue or the like (Wallsten, 1987) are time-consuming and expensive. The delivery systems for such devices have also suffered from problems relating to the repositionability of the devices as they are delivered into position in the living creature.
In stenting long arterial segments, the contiguously decreasing diameter of the arterial system from the center to the periphery may pose problems. Woven stents with a uniform diameter will exert a substantial expansile force to the vessel wall along the tapered portion. Additionally, the stent may remain more elongated in the tapered portion. In a study where WALLSTENTs with a uniform diameter were used to bridge central venous obstruction in hemodialysis patients, it was found that the stents which were selected according to the size of the larger diameter central vein exerted considerably higher force to the wall of the smaller caliber subclavian vein (Vesely, 1997). Simultaneously, the length of the stents in the smaller caliber vein was longer than expected.
In the prior art, most of the filter designs except for the Bird's Nest filter (Cook Inc., Bloomington, Ind.) have a conical shape and are anchored with multiple legs in the wall of the cava. The conical design is used because the main stream of the blood carries the thrombi from the lower part of the body through the center of the inferior vena cava. Therefore, all these devices are designed to have good filtration capacity at the center of the cava. The situation is quite different after some thrombi have been successfully captured. The center of the cava will no longer be patent and as a result, the blood will be diverted from the center to the periphery of the cava. The aforementioned designs, however, are not capable of catching thrombi effectively at the periphery of the lumen so the patients will practically be unprotected against subsequent peripheral embolization (Xian, 1995; Jaeger, 1998). Further, most of filters tend to be tilted in the cava which can deter their thrombus-capturing efficacy. Additionally, except for the Simon nitinol filter (C.R. Bard, New Jersey, N.J.) the aforementioned designs require a fairly large invasive delivery system of 10-F or larger.
The uniform caliber of cylindrical stents in the prior art used in the ureter, as well as the peristalsis arrested at the proximal end of the stent, has resulted in severe hyperlasia of the urothelium and eventually occlusion of the ureter.
Turning to occluders, percutaneous occlusion techniques have become indispensable tools in minimally invasive management of a wide range of pathological conditions. Use of permanent mechanical occlusion devices has been shown to be equivalent to that of surgical ligation. The Gianturco-Wallace stainless steel coil (Cook Inc., Bloomington, Ind.) has been the most widely used permanent, expandable intravascular occlusion device for transcatheter delivery (Gianturco et al., 1975).
Percutaneous coil embolization has been shown to be advantageous over traditional surgical procedures in treatment of life threatening hemorrhage due to trauma or obstetric emergencies (Schwartz et al., 1993; Teitelbaum et al., 1993; Selby Jr., 1992; Levey et al., 1991; Ben-Menachem et al., 1991; Vedantham et al., 1997). Furthermore, coils have been used alone or in combination with microvascular embolic agents for the treatment of vascular fistulas and malformations, tumors, and varices (Wallace et al., 1979; Hendrickx et al., 1995; Furuse et al., 1997; White et al., 1996; Sagara et al., 1998; Punekar et al., 1996). During the last few years, the transcatheter closure of the patent ductus arteriosus (PDA) with coils has become a frequently used technique (Hijazi and Geggel, 1994; Hijazi and Geggl, 1997).
Although coil type occlusion devices have shown at least a degree of utility, they have a number of drawbacks that could be significant in some applications. Intravascular stability of the coils has been shown to be highly dependent on proper matching of coil diameter with the diameter of the target vessel (Nancarrow et al., 1987), and with the exception of small vessels, a single coil rarely results in a stable occlusive thrombus (Hijazi and Geggel, 1994). Moreover, a long vascular segment is often obliterated because of the frequent need for multiple coils and the coils often remain elongated within the vessel because their unconstrained diameter is larger than the vascular lumen. Furthermore, delayed recanalization rates of 37%–57% have been reported in humans within 1–3 months after initially successful coil embolization (Sagara et al., 1998; O'Halpin et al., 1984; Schild et al., 1994).
These and other drawbacks have inspired modifications in the design and technique of coil embolization. Recently, detachable microcoils and macrocoils with controlled delivery have been designed to achieve a more compact conglomerate of the coil and to prevent migration by allowing optimal positioning of the coil before release (Zubillaga et al., 1994; Guglielmi et al., 1995; Marks et al., 1994; Reidy and Qureshi, 1996; Uzun et al., 1996; Tometzki et al., 1996; Dutton et al., 1995). However, since optimal arrangement of the coil alone may not prevent migration in some cases, such as high flow conditions or venous placement, a coil anchoring system has been devised (Kónya et al., 1998). Although an anchoring system may stabilize a coil conglomerate within the vasculature, significantly reducing or eliminating the possibility of coil migration, such a system may render the coil non-repositionable.
Several different non-coil devices have been designed to achieve a more stable, limited size plug with higher hemostatic efficiency particularly for transcatheter closure of larger vessels (Schmitz-Rode et al., 1993; Kato et al., 1997; Kónya et al., 1999) and PDAs (Pozza et al., 1995; Magal et al., 1989; Grifka et al., 1996). Recently, initial clinical experiences with a new self-expanding nitinol-mesh PDA occluder have been reported (Sharafuddin et al., 1996; Masura et al., 1998). A similar self-expanding, repositionable quadruple-disc device constructed of a braided nitinol mesh and polyester fibers has been reported to be superior to standard Gianturco coils in experimental occlusion of mid-size arteries (Sharaffuddin et al., 1996).
Although such non-coil devices may be repositionable, they too exhibit drawbacks. For instance, the quadruple-disc device is several centimeters long in an elongated fashion, making difficult to keep the superselective position of the catheter tip during deployment. The multiple rigid connections between the layers and the relative long and rigid connection between the occluder and the delivery cable further increase this drawback. Although the nitinol mesh-PDA occluder has demonstrated utility, its proper placement requires a proper match both in size and shape between the occluder and the lesion to be occluded. The type and quality of the connection between the occluder and the delivery cable is the same as in the quadruple-disc design. A common disadvantage of both designs is that they lack guidewire compatibility. As a result, a delivery catheter must often be navigated to the site of occlusion first before an occluder may be loaded into the catheter and delivered through it. Another relative disadvantage of both devices is their cost of manufacturing.
Percutaneous catheter technique for permanent closure of isolated persistently patent ductus arteriosus (PDA) is now a treatment of choice among doctors, obviating open surgery. The configuration of the PDA varies considerably. A majority of PDAs tend to have a funnel or conical shape due to ductal smooth muscle constriction at the pulmonary artery insertion, although narrowings in the middle or aortic ends can be observed (Krichenko, 1989). That is the reason why not only the size, but also the configuration, of the lesion plays a significant role in selecting an appropriate occluding device. Except from the small caliber lesions (with a maximum diameter of 2.5 mm or 3.3 mm, respectively), where some authors have achieved successful closure of the PDA with Gianturco coils (Cambier, 1992; Lloyd, 1993; Sommer, 1994), Rashkind's “double umbrella” occluder is the most often used device for this purpose (Rashkind, 1987; Hosking, 1991; Latson, 1991; Wessel, 1988; Report of the European Registry, 1992). It is available in two sizes (with a diameter of 12 mm and 17 mm) which require a 8-F and 11-F delivery system, respectively.
In the majority of cases, the deployment of the traditional PDA device is performed from a femoral vein access (Report of the European Registry, 1992). Because of the size of the delivery sheath, such a device is not suitable for the treatment of patients with a body weight of less than 8 kg. Using even a larger umbrella, this procedure is not recommended for the treatment of the lesions with a diameter of 8 mm or above (Latson, 1991). About 80% of unselected patients with isolated PDA are candidates for the Rashkind device using the aforementioned criteria (Latson, 1991). With the Rashkind device, the proportion of patients with residual flow through the lesion fell from 76% immediately after implantation to 47% by the day after implantation and to 17% by a year after implantation (Report of the European Registry, 1992). According to some authors the residual flow carries a potential risk of infective endocarditis and should be avoided if possible. Its abolishment can be achieved by implantation of another device or surgery.
One of the main drawbacks of the Rashkind umbrella is that it is not suitable for occlusion of all types of PDA. Preferably, it is used to occlude short PDAs with relatively wide end-openings. Its two discs cover both the pulmonary and the aortic opening of the PDA. Longer PDA may hinder the discs to be positioned in the proper way, that is, parallel to each other, thereby deteriorating its self-anchoring. Another disadvantage of the umbrella is that the occluding capacity of the design depends exclusively on the thrombogenicity of the porous Dacron material, frequently resulting in partial and lengthy occlusion.
For the majority of patients with urinary leakage and/or fistulas (mainly due to tumor propagation to their ureters), the diversion of urine is currently performed by a percutaneous transrenal approach together with ureteral occlusion. Formerly, detachable and non detachable balloons were used for this purpose, but they did not cause satisfactory ureteral occlusion. Migration as well as deflation of the balloons occurred relatively frequently (Gunter, 1984; Papanicolau, 1985) leading to recurrence of the urine leakage. A silicone ureteral occluder was developed and used with only limited success because of device migration (Sanchez, 1988). This resulted in repositioning and consequent incomplete ureteral occlusion. It appears that the best results have been accomplished with Gianturco coils and Gelfoam embolization (Gaylord, 1989; Bing, 1992 a; Farrel, 1996). Even with multiple coil placements, together with Gelfoam plugs, the ureteral occlusion may sometimes be achieved for only weeks or months, and was attributed mostly to the induced urothelial hyperplasia (Bing, 1992 b). Coil migration was frequently encountered in these studies. The lack of appropriate self-anchoring results in coil migration which eventually deteriorates the occlusive effect.
Problems pointed out in the foregoing are not intended to be exhaustive but rather are among many that tend to impair the effectiveness of previously known stents, occluders and filters. Other noteworthy problems may also exist; however, those presented above should be sufficient to demonstrate that previous techniques appearing in the art have not been altogether satisfactory, particularly in providing flexible, self-expanding, repositionable stents, occluders and filters.